AWS · Certification path · MLA-C02
MLA-C02: Machine Learning Engineer - Associate
40 stages in study order, about 16 hours to complete, drawn from AI and machine learning, Data engineering and Security. Each stage names the section of the MLA-C02 exam guide it practises.
The path, in study order
- 1Right ToolFreeAI and machine learning · AI World 1 · Skill 2.1.3, Skill 2.1.4, Skill 2.1.8
- 2Pick the ModelAI and machine learning · AI World 1 · Skill 2.1.1, Skill 2.1.7, Skill 3.1.4, Skill 3.1.5, Skill 3.2.1, Skill 4.2.7
- 3Shape the AnswerAI and machine learning · AI World 1 · Skill 2.2.8, Skill 3.3.7
- 4Teach and MeasureAI and machine learning · AI World 1 · Skill 1.2.9, Skill 2.1.2, Skill 2.1.5, Skill 2.3.1, Skill 2.3.6, Skill 2.3.7, Skill 2.3.8, Skill 2.3.9, Skill 2.3.10
- 5Guard the GateAI and machine learning · AI World 1 · Skill 4.3.4, Skill 4.3.9
- 6Govern ItAI and machine learning · AI World 1 · Skill 1.3.3, Skill 2.3.4, Skill 4.3.2, Skill 4.3.3, Skill 4.3.4, Skill 4.3.8
- 7Feed the BaseAI and machine learning · AI World 2 · Skill 3.2.7, Skill 3.3.11
- 8Cut and EmbedAI and machine learning · AI World 2 · Skill 1.2.5, Skill 1.2.7, Skill 2.2.9
- 9Pick the StoreAI and machine learning · AI World 2 · Skill 1.1.7, Skill 2.1.5, Skill 3.2.7
- 10Find the Right ChunkAI and machine learning · AI World 2 · Skill 3.1.8, Skill 3.2.8
- 11Prove the RetrievalAI and machine learning · AI World 2 · Skill 2.3.1, Skill 2.3.9, Skill 2.3.10
- 12Nothing FoundAI and machine learning · AI World 2 · Skill 2.3.10, Skill 3.3.11
- 13Give It ToolsAI and machine learning · AI World 3 · Skill 3.1.6
- 14Run It SomewhereAI and machine learning · AI World 3 · Skill 3.1.6, Skill 3.2.11, Skill 3.3.8
- 15Remember MeAI and machine learning · AI World 3 · Skill 3.2.9
- 16Know When to StopAI and machine learning · AI World 3 · Skill 4.3.2, Skill 4.3.3
- 17Guard the AgentAI and machine learning · AI World 3 · Skill 4.3.9
- 18Read the TraceAI and machine learning · AI World 3 · Skill 4.1.1, Skill 4.1.5, Skill 4.2.2
- 19Shape the DataAI and machine learning · AI World 4 · Skill 1.1.2, Skill 1.1.5, Skill 1.1.9, Skill 1.2.1, Skill 1.2.2, Skill 1.3.1, Skill 1.3.7
- 20Label ItAI and machine learning · AI World 4 · Skill 1.3.2, Skill 2.3.7
- 21Train and TuneAI and machine learning · AI World 4 · Skill 2.1.6, Skill 2.2.1, Skill 2.2.2, Skill 2.2.3, Skill 2.2.4, Skill 2.2.7, Skill 2.3.1
- 22Check the BiasAI and machine learning · AI World 4 · Skill 1.3.3, Skill 1.3.5, Skill 2.3.4
- 23Read the NumbersAI and machine learning · AI World 4 · Skill 2.2.5, Skill 2.3.6
- 24Wire the PipelineAI and machine learning · AI World 4 · Skill 3.3.1, Skill 3.3.3, Skill 3.3.4, Skill 3.3.5, Skill 3.3.6
- 25Pick the EndpointAI and machine learning · AI World 5 · Skill 3.1.1, Skill 3.1.3, Skill 3.1.7
- 26Scale ItAI and machine learning · AI World 5 · Skill 3.2.6, Skill 3.2.10, Skill 4.2.1, Skill 4.2.4
- 27Keep It PrivateAI and machine learning · AI World 5 · Skill 3.2.4, Skill 4.3.6
- 28Lock It DownAI and machine learning · AI World 5 · Skill 4.3.2, Skill 4.3.3, Skill 4.3.4, Skill 4.3.5
- 29Test in ProductionAI and machine learning · AI World 5 · Skill 2.3.3, Skill 3.3.1, Skill 4.1.4
- 30Watch the DriftAI and machine learning · AI World 5 · Skill 2.3.2, Skill 4.1.2, Skill 4.1.3, Skill 4.2.3, Skill 4.2.5, Skill 4.2.10
- 31Shape the FilesFreeData engineering · Data World 1 · Skill 1.1.5
- 32Pick the StreamData engineering · Data World 3 · Skill 1.1.4
- 33Hot ShardData engineering · Data World 3 · Skill 1.1.3
- 34Many ReadersData engineering · Data World 3 · Skill 1.1.3
- 35Land It in S3Data engineering · Data World 3 · Skill 1.1.5
- 36Pick the StoreData engineering · Data World 5 · Skill 1.1.2, Skill 1.1.7
- 37Quality GateData engineering · Data World 6 · Skill 1.3.1
- 38Tag Says SoSecurity · Security World 1 · Skill 4.3.3
- 39Who Are YouSecurity · Security World 1 · Skill 4.3.5
- 40Hardened HostsSecurity · Security World 4 · Skill 4.3.1
Articles on MLA-C02
MLA-C02 · AI and machine learning
Amazon Bedrock or SageMaker AI: the question that decides
Bedrock serves foundation models through an API. SageMaker AI builds, trains and hosts your own. The deciding question is where the answer has to come from.
Ends at Right ToolFree5 min read · video coming
AIF-C01 · AI and machine learning
Which AWS AI service for which request: read the input, then the output
Audio, scans, photos, text or rows with known outcomes. The form of the input and the output asked for pick the AWS AI service, and rule out the near misses.
Ends at Right ToolFree5 min read · video coming
DEA-C01 · Data engineering
Why Athena reads 1.2 TB to answer a four-column query
Athena bills the bytes it scans. Text is read whole; Parquet and ORC read only the columns a query names. How format, compression and file size set the bill.
Ends at Shape the FilesFree5 min read · video coming
Guide sections the path practises
| Section | As the guide words it | Stages |
|---|---|---|
| Skill 1.1.2 | Make storage decisions and configure storage services based on cost, performance, data structure, and data compliance | 2 |
| Skill 1.1.3 | Troubleshoot and debug data ingestion and storage issues that involve capacity and scalability | 2 |
| Skill 1.1.4 | Use AWS streaming data sources to ingest data (for example, by using Amazon Kinesis, Apache Flink, Apache Kafka) | 1 |
| Skill 1.1.5 | Ingest from and write by using appropriate data formats (for example, Apache Parquet, JSON, CSV, ORC) based on data access patterns | 3 |
| Skill 1.1.7 | Configure scalable vector databases for AI applications (for example, OpenSearch Service, Amazon RDS with pgvector, Amazon S3) based on specifications | 2 |
| Skill 1.1.9 | Ingest data into SageMaker Feature Store | 1 |
| Skill 1.2.1 | Transform data by using AWS tools (for example, AWS Glue, AWS Glue DataBrew, Spark on Amazon EMR, SageMaker Data Wrangler) | 1 |
| Skill 1.2.2 | Create and manage features by using AWS tools (for example, SageMaker Feature Store) | 1 |
| Skill 1.2.5 | Configure and use embedding models to transform text and image data into numerical representations | 1 |
| Skill 1.2.7 | Prepare documents for Retrieval Augmented Generation (RAG) applications (for example, chunking strategies, metadata extraction) | 1 |
| Skill 1.2.9 | Prepare data for FM fine-tuning, continuous pre-training, and model distillation | 1 |
| Skill 1.3.1 | Validate data quality (for example, by using DataBrew, AWS Glue Data Quality) | 2 |
| Skill 1.3.2 | Label and annotate data | 1 |
| Skill 1.3.3 | Identify and mitigate sources of bias in data by using AWS tools and techniques (for example, dataset splitting, shuffling, augmentation) | 2 |
| Skill 1.3.5 | Resolve class imbalance in numeric, text, and image datasets | 1 |
| Skill 1.3.7 | Clean data (for example, by detecting outliers, imputing missing data, deduplication) | 1 |
| Skill 2.1.1 | Evaluate and select appropriate FMs from Amazon Bedrock based on task requirements and performance criteria | 1 |
| Skill 2.1.2 | Identify fine-tuning strategies for pre-trained FMs to meet business needs | 1 |
| Skill 2.1.3 | Compare and select appropriate ML models, generative AI (GenAI) models, algorithms, and solution templates to meet business needs (for example, interpretability, domain-specific performance, latency) | 1 |
| Skill 2.1.4 | Evaluate tradeoffs between custom solutions, managed services, pre-trained models, and FMs to meet business needs | 1 |
| Skill 2.1.5 | Select Retrieval Augmented Generation (RAG) architecture patterns based on use case requirements | 2 |
| Skill 2.1.6 | Assess tradeoffs between ML model performance, training time, and cost | 1 |
| Skill 2.1.7 | Assess tradeoffs between AI model performance, latency, and cost | 1 |
| Skill 2.1.8 | Apply AWS AI services to solve specific business problems (for example, Amazon Textract, Amazon Rekognition, Amazon Comprehend, Amazon Transcribe) | 1 |
| Skill 2.2.1 | Apply Amazon SageMaker AI built-in algorithms and common ML libraries | 1 |
| Skill 2.2.2 | Configure SageMaker AI script mode with supported frameworks for simplicity and performance | 1 |
| Skill 2.2.3 | Implement hyperparameter optimization (for example, SageMaker AI automatic model tuning [AMT]) | 1 |
| Skill 2.2.4 | Implement training time reduction techniques (for example, early stopping, distributed training) | 1 |
| Skill 2.2.5 | Prevent model overfitting, underfitting, and catastrophic forgetting | 1 |
| Skill 2.2.7 | Adjust fundamental hyperparameters (for example, epoch, steps, batch size) | 1 |
| Skill 2.2.8 | Apply customization techniques for AI solutions (for example, task-specific prompt engineering, fine-tuning) | 1 |
| Skill 2.2.9 | Optimize retrieval components and embedding models. | 1 |
| Skill 2.3.1 | Perform reproducible experiments (for example, by using MLflow on SageMaker AI, Amazon Bedrock evaluations, Amazon Bedrock Prompt Management) | 3 |
| Skill 2.3.2 | Create model performance baselines and implement drift detection | 1 |
| Skill 2.3.3 | Compare the performance of shadow variants to production variants | 1 |
| Skill 2.3.4 | Explain model outputs | 2 |
| Skill 2.3.6 | Apply comprehensive model evaluation techniques for traditional ML and GenAI models | 2 |
| Skill 2.3.7 | Implement integrated human evaluation frameworks (for example, human-in-the-loop workflows, text generation quality assessment) | 2 |
| Skill 2.3.8 | Apply natural language processing (NLP) evaluation metrics (for example, bilingual evaluation understudy [BLEU], Recall-Oriented Understudy for Gisting Evaluation [ROUGE], BERTScore, semantic similarity) | 1 |
| Skill 2.3.9 | Perform AI evaluation (for example, model output assessment, content quality validation, bias detection, LLM-as-a-judge frameworks) | 2 |
| Skill 2.3.10 | Configure RAG system monitoring, including retrieval accuracy assessment | 3 |
| Skill 3.1.1 | Select appropriate compute environments and deployment targets | 1 |
| Skill 3.1.3 | Select model inference strategies (for example, real-time and batch processing) | 1 |
| Skill 3.1.4 | Evaluate and select appropriate foundation model (FM) deployment options | 1 |
| Skill 3.1.5 | Deploy models that were built outside of AWS into AWS environments (for example, Amazon SageMaker AI, Amazon Bedrock Custom Model Import) | 1 |
| Skill 3.1.6 | Deploy and configure agents for specific tasks, integration with other services and tools, and agent communication protocols | 2 |
| Skill 3.1.7 | Configure FM deployment, model hosting, and resource allocation | 1 |
| Skill 3.1.8 | Apply Retrieval Augmented Generation (RAG) system configurations (for example, retrieval strategies, reranking) | 1 |
| Skill 3.2.1 | Optimize resource provisioning between on-demand and provisioned resources for performance and cost efficiency | 1 |
| Skill 3.2.4 | Configure SageMaker AI endpoints within VPC network environments | 1 |
| Skill 3.2.6 | Select specific metrics for auto scaling implementations | 1 |
| Skill 3.2.7 | Create and manage Amazon Bedrock knowledge bases with vector database configurations, document indexing, and retrieval optimization | 2 |
| Skill 3.2.8 | Implement retrieval pipelines to meet business needs | 1 |
| Skill 3.2.9 | Implement agent state management systems | 1 |
| Skill 3.2.10 | Implement AI-specific resource scaling for GPU workloads | 1 |
| Skill 3.2.11 | Deploy agentic workflow infrastructure | 1 |
| Skill 3.3.1 | Implement automated deployment strategies and rollback actions | 2 |
| Skill 3.3.3 | Configure training and inference jobs | 1 |
| Skill 3.3.4 | Configure automated testing strategies within CI/CD pipelines for traditional ML and AI workloads | 1 |
| Skill 3.3.5 | Build and integrate mechanisms to re-train models | 1 |
| Skill 3.3.6 | Manage model versions for repeatability and audits (for example, SageMaker Model Registry, MLflow on SageMaker AI) | 1 |
| Skill 3.3.7 | Manage prompts (for example, Amazon Bedrock Prompt Management) | 1 |
| Skill 3.3.8 | Implement automated agent deployment pipelines and agent version management | 1 |
| Skill 3.3.11 | Configure AI-specific pipeline orchestration for RAG system updates and knowledge base refresh cycles | 2 |
| Skill 4.1.1 | Monitor model performance in production by using Amazon CloudWatch generative AI observability, Amazon Bedrock Model Evaluation, and drift detection pipelines | 1 |
| Skill 4.1.2 | Monitor workflows to detect anomalies or errors in data processing or model inference | 1 |
| Skill 4.1.3 | Detect changes in data distribution that can affect model performance | 1 |
| Skill 4.1.4 | Monitor model performance in production by using A/B testing | 1 |
| Skill 4.1.5 | Monitor and automate the management of agent performance and coordination (for example, coordination failure detection, truncated streaming, tool failures) | 1 |
| Skill 4.2.1 | Select inference instance families to optimize performance and cost | 1 |
| Skill 4.2.2 | Configure and use tools to troubleshoot and analyze resources (for example, Amazon CloudWatch, Amazon Bedrock AgentCore Observability, AWS X-Ray) | 1 |
| Skill 4.2.3 | Set up dashboards to monitor performance metrics | 1 |
| Skill 4.2.4 | Optimize capacity for cost, performance, and reliability | 1 |
| Skill 4.2.5 | Optimize costs and set cost quotas by using appropriate cost management tools | 1 |
| Skill 4.2.7 | Evaluate cost implications of using FMs for inference in production | 1 |
| Skill 4.2.10 | Monitor AI-specific cost patterns (for example, token usage optimization, embedding computation costs, vector database storage optimization) | 1 |
| Skill 4.3.1 | Secure continuous integration and continuous delivery (CI/CD) pipelines by checking for code and image vulnerabilities (for example, by using Amazon CodeGuru, Amazon Inspector) | 1 |
| Skill 4.3.2 | Configure least privilege access to ML and AI artifacts | 3 |
| Skill 4.3.3 | Configure IAM policies and roles for users and applications in ML and AI systems | 4 |
| Skill 4.3.4 | Configure comprehensive monitoring, auditing, compliance, and logging for ML and AI systems (for example, AWS CloudTrail, AWS Config) | 3 |
| Skill 4.3.5 | Troubleshoot and debug security issues in ML and AI systems | 2 |
| Skill 4.3.6 | Create VPCs, subnets, and security groups to securely isolate ML and AI systems | 1 |
| Skill 4.3.8 | Select the appropriate credential type to access FMs (for example, Amazon Bedrock API keys, IAM credentials) | 1 |
| Skill 4.3.9 | Implement safeguards and sensitive data protection to meet application requirements and responsible AI policies (for example, by using Amazon Bedrock Guardrails) | 2 |
A domain and a path are separate products. What you own is the union of what you bought, so a stage you already own is never sold to you twice: the game offers the cheapest way to unlock the stage in front of you. Path $15, exam $150.